This review was researched and written by ACLU legal intern Max Bauer
In recent years, video surveillance technology has become a huge presence in both the United Kingdom and the United States. In the United States, following the September 11, 2001 attacks, governments have spent billions on surveillance cameras to prevent crime while privacy advocates have warned about the erosion of personal liberty. Balancing these two competing interests – privacy and security – requires a careful analysis of how effective cameras are at reducing crime and at what cost, both financial and constitutional. This paper reviews the literature on the effectiveness of camera surveillance systems and responses to such research, and measures the costs and benefits of these systems.
Earlier research in the UK (pre-21st century)
The United Kingdom experienced a sharp increase in the presence of closed-circuit television (CCTV) surveillance cameras starting in the 1980s. Despite this proliferation of cameras, the country had no significant drop-off in crime at the end of the century. The research commissioned by the British Home Office, the agency responsible for security in the United Kingdom, subsequently provided two of the early scientific studies on the effects of CCTV surveillance cameras in the early 21st century. These studies, described below, showed no significant impact in crime due to the presence of cameras.
British Home Office Study, 2002
The first Home Office study in August 2002, conducted by Brandon Welsh and David Farrington, surveyed 22 studies of CCTV in both the US and UK for a meta-analysis and found that as a whole the cameras showed no significant impact on crime. Welsh and Farrington’s data showed a very small impact on crime that was statistically insignificant. In studies included in their meta-analysis which did show a reduction in crime other interventions, such as improved lighting, fencing, notices about CCTV and increased security personnel confounded the data such that any reduction could be solely attributable to the cameras themselves.
The authors (along with additional co-authors) asserted this opinion in a subsequent paper: “Overall, it might be concluded that CCTV reduces crime to a small degree.”
British Home Office Study, 2005
A February 2005 meta-analysis of fourteen sites commissioned by the Home Office and conducted by Martin Gill and Angela Spriggs reached a similar conclusion to that of Welsh and Farrington. Only one site they studied experienced a statistically significant crime reduction and this result was limited by confounds. Furthermore this was the most expensive site. This study also showed an increase in reported crime in half the sites which the authors hypothesize was due to increased awareness of crime. Spriggs and Gill concluded: “It has been shown that the CCTV schemes produced no overall effect on all relevant crime viewed collectively.”
Within this meta-analysis however, the authors did find a reduction in vehicle crimes in half the sites though it was only statistically significant in two such sites although a change in parking regulations reduced the number of cars at the site and therefore opportunities for vehicle crimes. Additionally, violent crimes against individuals actually increased in three of four urban sites, which was similar to crime patterns in the control. Gill and Spriggs cautioned: “The belief at that CCTV alone can counter complex social problems is unrealistic in the extreme.” However, they also warned against concluding CCTV was not effective based on their study. A later report by the British House of Lords noted public opinion is important as a policy consideration and warned against concluding that people feel safer merely due to the presence of cameras.
San Francisco – UC Berkeley, CITRIS Report, 2008
In the United States, the first major scientific surveillance camera study was conducted based on San Francisco’s Community Safety Camera (CSC) program. Jennifer King of University of California (UC) Berkeley Law and her colleagues conducted a six-month study for San Francisco through the University’s Center for Informational Research in the Interest of Society (CITRIS) releasing a final report December 17, 2008. The study found very little impact on violent crime and that homicides did decline near the cameras but increased further away (a displacement effect but not reduction). The study did find a statistically significant decrease in property crime near the cameras. Finally there was little of evidence of an impact on other types of crimes in the vicinity of the cameras.
The report concluded that at least for property crimes, “the system is clearly having an effect” but noted that the CSC program’s “lack of deterrent effects on violent crime and its limited usefulness with respect to investigations” limited the program’s benefit. The camera program was incredibly expensive, especially given its failure to reduce violent crime. Finally, because the study was short and featured only one city, its results could be a fluke.
Los Angeles – USC Study, 2008
Students at the University of Southern California (USC) School of Policy released a report for the California Research Bureau in May 2008 on the effects of video surveillance cameras in two areas of Los Angeles. The study showed no impact on crime based on the Los Angeles Police Department’s crime statistics before and after the camera installation. While violent crime declined in both areas this trend merely followed the general pattern also demonstrated in the control areas. The study found no statistically significant evidence of displacement.
Baltimore, Chicago & Washington, D.C. – Urban Institute, 2011
In September 2011, the Urban Institute published a study, primarily funded by the United States Department of Justice, looking at surveillance systems in three major American cities: Baltimore, Chicago & Washington, D.C. The authors concluded the findings from Baltimore “create a strong case for the positive impact that Baltimore’s cameras have on crime, with some
signs of diffusion of benefits and no statistically significant evidence of displacement.” However, this lack of displacement may simply indicate a general decline in crime independent from implementation of the camera program. They also noted the great costs of the system. An additional unintended externality the implementation caused was turf battles between rival gangs looking for space outside of the cameras’ views.
In Chicago, the study showed a reduction in crime in one neighborhood (Humboldt Park) but not in another (West Garfield Park). In a 2013 article, American Civil Liberties Union (ACLU) of Illinois attorney Adam Schwartz criticized the results by asserting the study did not control for the gentrification which occurred Humboldt Park during the period the study was conducted nor did it control for other police initiatives. As with Baltimore, the study noted the high cost of the camera network.
The Washington component of the study noted crime was on the rise in the summer of 2006 which prompted the installation of the cameras that fall but that in the next two summers crime rates matched pre-implementation levels. The overall crimes rates in the experimental and control areas were roughly the same. The chapter of this report on Washington, D.C., did not include a cost-benefit analysis.
In the fall of 2006 the New York Civil Liberties (NYCLU) published a report highlighting the increase in video surveillance and the need for oversight. This report noted the chilling effect cameras have on free speech and association and the threat to privacy and potential for abuse, including a racial disparate impact and specific risk of harassment invasive surveillance poses to individuals wearing skirts and blouses.
California ACLU Affiliates, 2007
On the opposite coast but in a similar vein, in August 2007 the ACLU affiliates in California published a report on the growing use of surveillance cameras in that state and these systems’ effect on civil liberties. The report noted the threats the cameras posed to the rights of freedom of speech, association and movement, and to privacy. Furthermore the report stressed that studies have shown cameras to be ineffective at deterring crime or solving crimes nor did they reduce fear of crime.
Chicago – ACLU of Illinois, 2011
In February 2011, the ACLU of Illinois published a large-scale report on Chicago’s network of video surveillance cameras. The ACLU’s Schwartz estimated the city has access to somewhere between 10,000 and 20,000 publicly and privately-owned cameras though the exact number is unknown. Like the NYCLU’s report and ACLU’s California report, this report did not scientifically examine the effectiveness of the surveillance camera system as a tool to fight crime but rather analyzed the system with respect to civil liberties. The study noted the following risks inherent in such a surveillance scheme as demonstrated by various surveillance camera systems: the absence of regulation of many of the cameras’ features, privacy and First Amendment problems, improper release of video by employees, and racial disparities in targeting. The study looked at the exorbitant costs of the cameras while noting such systems are not particularly effective for solving crimes and questioning their effectiveness in deterring crime.
Cost-Benefit Analysis Summary
As noted above, surveillance camera systems are excessively expensive, especially given their lack of effectiveness in both preventing and solving crime and their high risk of curtailing numerous civil liberties. The British Home Office for example spent 78% of its budget on cameras back in the 1990s while the United States Department of Homeland Security had given out $23 billion in American taxpayer money by 2007. In specific major US cities the following financial estimates are available: at least $60 million for Chicago’s camera network as of 2013 and $6.8 for three years, $8.1 million for Baltimore as of April 2008 (for three years), and $81.5 million for New York City’s counterterrorism operations which includes a city-wide CCTV system as of 2006. Despite these exorbitant price tags, as noted above there is little statistical evidence that the cameras reduce crime and cameras don’t prevent terrorism or provide much forensic value to investigators.
Then there is also another cost: civil liberties. The ACLU chapters’ reports cited above all reached similar conclusions above how pervasive camera technology jeopardizes fundamental rights. Individuals have a constitutional right be free from unreasonable searches and the Supreme Court has defined a search as unreasonable in circumstances where “a person has a constitutionally protected reasonable expectation of privacy.” The pervasive use of cameras, particularly when combined with advanced analytics and biometric tracking may change individuals’ expectations of privacy but when the Court established their test in the 1960s they could not foresee a time when all expectations of privacy would be eliminated due to constant video surveillance. Privacy interests may further be threatened as surveillance cameras are increasingly able to acquire biometric data.
Even if the Fourth Amendment protection is not triggered by simple video monitoring in public places, the First Amendment could be. Fear of surveillance can also implicate First Amendment freedoms of speech and association and freedom to assemble by chilling speech and activity. These rights are as fundamental as any rights to American democracy.
Finally, Justice Sotomayor’s concurring opinion in US v. Jones hints that the court may change its views on persistent, long-term government tracking of people in public places, as the technology advances and the cost of monitoring entire populations shrinks.
Notably, no scientific study has been conducted on the effectiveness of New York City’s surveillance camera scheme. The New York Police Department’s statistics on its stop-and-frisk policy show a disparate impact if not unconstitutional racial profiling. The use of video surveillance provides another opportunity for unequal treatment of citizens.
For example, some opponents of cameras argue cameras don’t eliminate but merely displace crime. See generally Sam Waples, e.al., Does CCTV Displace Crime?, 9 Criminology & Crim. Just. 207 (2009), abstract.
Nick Taylor, Closed Circuit Television: The British Experience, 1999 Stan. Tech. L. Rev. 11, 1 (1999).
B. Welsh & D. Farrington, “Crime Prevention Effects of Closed Circuit Television: A Systemic Review”, Home Office Research, Development and Statistics Directorate (2002); see Noam Biale, “Expert Findings on Surveillence Cameras: What Criminologists and Others Studying Surveillence Cameras Have Found”, ACLU at 3 (White Paper based on major studies 2000-2008), available here.
David Weisburd et. al., Effects of Closed-Circuit Television on Crime, 587 Annals Am. Acad. Pol. & Soc. Sci. 110, 133 (2003).
One of these sites had to be removed from the data set for failing to meet crime recording criteria so only thirteen sites were usable for scientific study.
Gill & Spriggs, “Assessing the Impact of CCTV”, Home Office Research, Development and Statistics Directorate (2005); see Biale, supra note 7, at 3-4.
Id. at 29-30.
Id. at 30.
Id. at 30.
Id. at 35.
Id. at 37.
Id. at 36.
Id. at 115; compare Nic Groombridge, “Stars of CCTV?: How the Home Office wasted millions – a radical ‘Treasury/Audit Commission’ view”, Surveillance Studies Network, at 76-77 (2008).
Surveillance: Citizens and the State, House of Lords, Select Committee on the Constitution, 2nd Report of Session 2008–09, at 92 (2009).
Jennifer King, et al., CITRIS Report: The San Francisco Community Safety Camera Program: An Evaluation of Effectiveness of San Francisco’s Community Safety Cameras, UC Berkeley, CITRIS (2008)(hereinafter CITRIS).
Id. at 84.
Id. at 156.
See Biale, supra note 7, at 5 (commentary on preliminary report).
Id. at 34, 37.
Id. at 4, 34, 38.
 See id. 44, 47.
Id. at 47-48.
Urban Institute study, at x, 64, 66.
Urban Institute study, at 69.
Id, at 79. This study does not count the Department of Homeland Security-funded cameras which were installed in the city soon after the September 11, 2001 attacks. Id. at 73.
Id. at 81, 83.
Id. at 9-10.
Id. at 10-11.
Id. at 11-12.
Mark Schlosberg & Nicole A. Ozer, Under the Watchful Eye: The Proliferation of Video Surveillence Systems in California, Cal. ACLU Affiliates (2007) (hereinafter Cal. ACLU).
Id. at 13.
Id. at 12.
Schwartz, supra note 37, at 9-12.
ACLU of Illinois, at 1, 7.
Id. at 14
Id., at 18
C. Norris & G. Armstrong, The Maximum Surveillance Society: The Rise of Closed Circuit Television at 54 (1999).
See Savage, supra note 3.
See Schwartz, supra note 37, at 31.
Urban Institute study, at 71.
Id., at 48, 50.
NYCLU, at 2.
U.S. Const. Amend IV
Katz v. United States, 389 U.S. 347, 360 (1967) (Harlan, J., concurring).
See Kyllo v. United States, 533 U.S. 27, 33-34 (2001) (“It would be foolish to contend that the degree of privacy secured to citizens by the Fourth Amendment has been entirely unaffected by the advance of technology.”); United States v. Jones, 132 S. Ct. 945, 963 (2012) (Sotomayor, J., concurring) (“The availability and use of these and other new devices will continue to shape the average person’s expectations about the privacy of his or her daily movements.”).
See Samuel D. Warren & Louis D. Brandeis, The Right to Privacy, 4 Harv. L. Rev. 193, 195, 196 (1890)( recognizing “instantaneous photographs… have invaded the secret precincts of private and domestic life…Of the desirability – indeed of the necessity – of some such protection there can, it is believed, be no doubt.”
See Sabrina A. Lochner, Note, Saving Face: Regulating Law Enforcement’s Use of Mobile Facial Recognition Technology & Iris Scans, 55 Ariz. L. Rev. 201, 208 (2013); Lucas D. Introna & Helen Nissenbaum, Facial Recognition Technology: A Survey of Policy and Implementation Issues (2009), 11, 15-17, available here.
Aileen B. Xenakis, Washington and CCTV: It’s 2010, Not Nineteen Eighty-Four, 42 Case W. Res. J. Int’l L. 573, 583 (2010) (arguing that under the Katz standard, “CCTV programs do not implicate the Fourth Amendment, as no person moving about public spaces has a reasonable expectation of privacy”).
U.S. Const. Amend I.
See ACLU of Illinois, at 15, 19; New York Times Co. v. Sullivan, 376 U.S. 254, 300 (1964) (“the Court conclusively demonstrates the chilling effect of the Alabama libel laws on First Amendment freedoms.”); Virginia v. Am. Booksellers Ass’n, Inc., 484 U.S. 383, 393 (“self-censorship; a harm that can be realized even without an actual prosecution”)certified question answered sub nom. Com. v. Am. Booksellers Ass’n, Inc., 236 Va. 168 (1988).
See De Jonge v. State of Oregon, 299 U.S. 353, 365 (1937) (First Amendment rights “maintain the opportunity for free political discussion… Therein lies the security of the Republic, the very foundation of constitutional government.”); see also Akhil Reed Amar, The Bill of Rights As A Constitution, 100 Yale L.J. 1131, 1147 (1991) (“Fittingly, the classic First Amendment dissents of Holmes and Brandeis were themselves exercises of free speech by a minority inspired by the hope of persuading a future majority (of the Court, of course).”); Jason Mazzone, Unamendments, 90 Iowa L. Rev. 1747, 1750 (2005) (“Freedom of speech, for example, is sometimes said to be so essential to democracy that there can be no constitutional amendment that repeals the speech and press clauses of the First Amendment”).